Transform service work with AI agents

The appliance story shows what changes when autonomous agents become part of service delivery. The customer still needs a resolution, the case still holds the service history, and people still own decisions. Agents help the organization understand requests, complete defined case work, and turn resolved cases into better knowledge.

Distinguish Copilot from autonomous agents

Copilot assists a person during service work. It can summarize the appliance conversation, answer a question from approved knowledge, or draft a response for a representative to review. The representative remains in control of the next action.

Autonomous agents perform defined work on behalf of the organization. Their instructions, permissions, data, and escalation paths determine what they can do without waiting for a person at every step. In Dynamics 365 Customer Service, three agents support a connected cycle: understand the need, manage the case, and improve the knowledge used in future service.

Explore autonomous agents in Customer Service

Each agent focuses on a different service outcome. Together, they use the records and knowledge introduced earlier in this module.

Case Management Agent

Case Management Agent helps automate work across the case lifecycle. It can create and update cases, support resolution, follow up with customers, and close cases according to its configuration. This reduces the time representatives spend entering details and completing routine case actions.

In the appliance story, the agent can use the conversation context to help create a case with the customer, product, and issue details already collected. As the repair progresses, it can update the case and support follow-up before closure. The case remains the durable record of the issue, activities, service commitments, and outcome.

The organization defines when the agent acts and when work transfers to a representative. A person handles exceptions, verifies sensitive or uncertain information, and makes decisions that require judgment. For current usage details, see Use Case Management Agent to create and update cases.

Customer Intent Agent

Customer Intent Agent uses generative AI to identify common customer intents and suggest new ones. An intent describes what customers are trying to accomplish, such as troubleshoot an error, check an order, or schedule service. Intent differs from a conversation, which records an interaction, and a case, which tracks an issue through resolution.

For the appliance company, the agent can discover that many customers contact support about a washing machine error code. A defined intent helps the service experience ask relevant follow-up questions, present an appropriate solution, or route the request to a representative with the right skills. Intent data also helps the organization understand why customers seek help.

Administrators review suggested intents and decide how to use them in self-service, representative assistance, or intent-driven routing. For current usage details, see Use Customer Intent Agent-based suggestions to handle conversations.

Customer Knowledge Management Agent

Customer Knowledge Management Agent extracts knowledge from resolved cases and helps manage customer service knowledge. After a case closes, it analyzes the case details, identifies knowledge gaps, and helps create or update knowledge articles while supporting the organization's compliance requirements.

In the appliance story, several resolved cases might reveal that the troubleshooting article omits an important reset step. The agent can identify the gap and prepare knowledge for review. Once approved and published, the improved article supports the next customer, the next representative, and AI-assisted responses.

Knowledge review remains important because a resolved case can contain information that is incomplete, specific to one customer, or unsuitable for reuse. The organization controls article approval and publication. For current administration details, see Use Customer Knowledge Management Agent to create and manage knowledge articles.

Connect the agents across the service cycle

The three agents form a continuous transformation loop:

• Understand the need: Customer Intent Agent identifies why customers request help and supports the next appropriate action.
• Manage the work: Case Management Agent performs defined tasks as the case moves from creation through resolution and closure.
• Improve future service: Customer Knowledge Management Agent turns resolved-case patterns into knowledge improvements for review.

The output from one stage strengthens the next. Clear intent provides better context for service work. Complete case data supports useful knowledge. Better knowledge improves future self-service and assisted service. This cycle reduces repeated effort without removing human responsibility.

Human oversight remains essential throughout the cycle. Administrators define instructions, permissions, approved knowledge, and escalation paths. Representatives handle work that requires judgment, and supervisors monitor quality and outcomes. Organizations evaluate accuracy, security, privacy, compliance, and performance before expanding an agent's responsibilities.

AI doesn't replace the service lifecycle. It helps service teams understand needs, act on cases, and improve knowledge while connected data carries context through the journey. Availability depends on product, licensing, region, and configuration. For current capability details, see Agents, Copilot, and AI capabilities in Dynamics 365 apps.

The three Customer Service agents strengthen the work managed within the case lifecycle. Resolving the appliance issue might also require voice or digital engagement, an on-site repair, or a process designed for the organization. The next unit explores how other Dynamics 365 applications and Microsoft Power Platform extend the service experience.